Nailing the solar activity – global temperature divergence lie

Posted: July 21, 2010 by tallbloke in solar system dynamics

We are frequently told that the Sun can’t be responsible for late C20th warming because temperature has increased while solar activity has dropped from it’s peak in the 1950’s.

What a load of rubbish.

Solar cycle amplitudes are only part of the story. The cycles in the late C20th were short, ~10 years, and high compared to the long term average of ~40 SSN. The minima between them were short too. So although they did reduce in absolute amplitude after the ’50s, they made up for it by kicking out more energy more of the time. Last year to get a handle on this, I integrated the total sunspot areas as a running cumulative total departing from the long term average.


The Sea Surface Temperature graph from with the trend lines added shows how well the sunspot cumulative total works as a proxy for Ocean Heat Content. The SST data is smoothed over 1/3 of the solar cycle length to bring out the solar effect on SST’s. I further developed this idea  in my simple solar-planetary energy model. Looking at the flattening of the rise at cycle 19-20 (1954-1976) and from cycle 22-23 and now the low cycle 24, I would say we are just over the top of the warming curve. The slightly falling OHC data from 2003 onwards measured by ARGO would seem to back this up.

So although we have yet to understand all the mechanisms by which the Sun’s energies get transferred into Earth’s climate system, we can say that the solar data fits the temperature record better than co2 data does, over a longer period too.

  1. orkneylad says:

    most excellent….nice job.

  2. Gray says:

    Excellent tallbloke were you able to go back before 1880?

  3. tallbloke says:

    That’s where the sunspot area data runs out. If you look at the linked article, I used sunspot numbers which are ok, back to 1840.

  4. DirkH says:

    A rather beautiful correlation!
    In case you missed it, here’s Habibullo Abdassamatov about the coming LIA; he talks about solar cycles.

  5. tallbloke says:

    Thanks Dirk,
    I’ve been too busy making my calculator hot to catch up with these conference presentations. Keep ’em coming.

  6. Tenuc says:

    Good graph TB – the correlation is excellent! This provides a solid platform for further thoughts…

    The changing orbits of the planets act as a governor on the solar power source which drives the solar dynamo(s) accelerating and checking the speed of activity. This results in:-

    Bigger sunspot area = higher proportion of UV
    Bigger sunspot area = stronger solar toroidal field (also larger poloidal field for next cycle?)
    Bigger sunspot area = larger subsequent coronal holes
    Larger coronal holes = faster/denser/more ionised solar wind

    The above changes effect atmosphere and ocean circulation/polar ice cover/cloud type/tropical storms, which kick our climate regime into a new state.

    The small observed changes in TSI are not the main drivers of short term change.

    The mechanisms by which earth is affected by solar changes are still not known.

    Now have I omitted or misconstrued anything?

  7. DirkH says:

    Don’t know if this has been covered here. Every 8 minutes a magnetic portal between the fields of Sun and Earth opens and rolls across the winter pole:
    Flux transfer event
    According to Dr. Leif Svalgaard the energy transferred is 1/10000 of the energy of a hurricane; in Watt, in other words, for as long as the portal is open, 1/10000 of the wattage of a hurricane flows. (He said so in a recent WUWT comment)

  8. tallbloke says:

    It’s worth reproducing Alan’s graphic here Tenuc. Thanks
    Alan Cheetham - solar regime shifts

  9. GoodGrief says:

    Total crap!

    [Reply] Good lad, now wipe it properly and give me a valid email address next time.

  10. sphaerica says:

    What statistical method did you use to compute the trend lines in your first graph?

    Your description of computing the “Cumulative sunspot total…” is unclear to me. Can you write it mathematically, so that I can be sure that I understand how you are deriving the numbers?

  11. jorgekafkazar says:

    Yes, interesting. But “bump matching” is problematic, especially when there are so many bumps that don’t actually coincide. Can you derive some sort of correlation coefficient for this? And sunspot area might not be the best measure of solar output to use. You might try following the UV-only output instead of spots, or even TSI, and see what you get.

    [Reply:] Hi Jorge, the UV and TSI measurements haven’t been around long enough to be of much use, are contentious (see PMOD vs ACRIM debacle), and the proxies are based on geomagnetic data which is contaminated with multiple variables, so sunspot areas or numbers are as good as anything available. Neither of these quantities are easily converted to watts, and we don’t know the amplification levels due to effects on cloud albedo prior to 1980. Dandifying graphs with false precision in the face of great uncertainty is a waste of time as far as I can see, so wiggle matching will have to do for now.

  12. Jack Simmons says:

    GoodGrief, could you be a bit more specific?


    Really nice work. Gives me something to think about.

    The butterfly pattern of sunspot migration during the solar cycle is one of my favorite patterns in nature. Love to study it. I like the way you display it with temperatures.

    [reply:] Hi Jack, and thanks. The graph you refer to was done by Alan Cheetham over at appinsys (see blogroll left) so you can get more detail there.

  13. Philip T. Downman says:

    Graphs without y-scale and sort! That disturbs at least me.

    [reply:] See my reply to Jack above. Any problem with the Y axes on my graph in the headline post, let me know the specifics and I’ll try to help.

  14. […] Solar cycle amplitudes are only part of the story. The cycles in the late C20th were short, ~10 years, and high compared to the long term average of ~40 SSN. The minima between them were short too. So although they did reduce in absolute amplitude after the ’50s, they made up for it by kicking out more energy more of the time. Last year to get a handle on this, I integrated the total sunspot areas as a running cumulative total departing from the long term average…. (Source location) […]

  15. tallbloke says:

    Tenuc says:
    July 21, 2010 at 3:38 pm (Edit)

    Good graph TB – the correlation is excellent! This provides a solid platform for further thoughts…

    The changing orbits of the planets act as a governor on the solar power source which drives the solar dynamo(s) accelerating and checking the speed of activity. This results in:-

    Bigger sunspot area = higher proportion of UV
    Bigger sunspot area = stronger solar toroidal field (also larger poloidal field for next cycle?)
    Bigger sunspot area = larger subsequent coronal holes
    Larger coronal holes = faster/denser/more ionised solar wind

    The above changes effect atmosphere and ocean circulation/polar ice cover/cloud type/tropical storms, which kick our climate regime into a new state.

    The small observed changes in TSI are not the main drivers of short term change.

    The mechanisms by which earth is affected by solar changes are still not known.

    Now have I omitted or misconstrued anything?

    Only the unknown unknowns. 😉

  16. @Tallbloke

    Click to access abdusamatov2.pdf

    Changes related to diameter and brightness. K.Abdusamatov:
    We show that the cyclic 11-year variations observed in the solar integral radiation flux are due to some fundamental global processes which occur deeply in the Sun and which cause the corresponding changes in the radius and effective temperature of the photosphere. The 11-year variations of the “solar constant” are determined almost entirely by the changes of the area of the radiating photosphere surface at a constant effective temperature. So, an 11-year heliocycle is a simultaneous phase- and amplitude-correlated oscillation of the solar activity, radius, and radiation flux. The secular variations of the “solar constant” were detected directly for the first time. We suppose that the observed identical long-term variations of the activity, radius, and radiation flux are caused by some common processes which occur deeply in the Sun. They correlate with the global oscillations of the whole Sun. Such oscillations are caused by cyclic changes of the temperature in the solar core, and they can trigger the generation of the solar cycles. We expect that the next relatively deep minimum of the solar activity, radius, and radiation flux in the 200-year quasi-cycle will be close to the Maunder minimum level and will occur in the year 2040 ±10.
    Though if the Sun is powered from the outside it does no necessarily means a hot interior, the current being modulated by the planets. If we could “see” with a blink of an eye of, say, a thousand years, we could see a somehing like a Ruhmkorff coil, the Sun being the core.

  17. tallbloke says:

    Jim D says:
    July 24, 2010 at 7:22 am (Edit)

    tallbloke says:
    July 24, 2010 at 1:21 am

    Back radiation can’t heat the oceans.

    Yes it can. There are hundreds of W/m2 of back radiation at the surface. Where does it go? Any respectable ocean energy budget has to include this, otherwise the temperature would drop towards the now famous 255 K.

    As Pileke pere correctly stated, the greenhouse effect works by slowing the rate of cooling, not by heating anything up. Co2 emitted radiation is long wave, and can’t penetrate the ocean much beyond its own wavelength. This is known physics. There is all sorts of worthless speculation about conduction from the ‘skin’ of the ocean, but take it from me, the sun heats the ocean because its radiation is mostly shortwave, which penetrates to several tens of metres into the briny. Longwave radiation just causes more evaporation at the surface. Any conduction effects are second order.

    It was put to the test by the realclimate alarmists, who were *desperate* to show some effect from back radiation. They got something like a 0.0002C change from cloudy to clear sky, and said that it ‘proved the principle’. What a joke. I laughed my a*se off.

    Now you might think that more evaporation at the surface might cause the secondary effect on water vapour and humidity levels the alarmists are needing to bump up the sensitivity, but you’d be wrong. See the post by ninderthana on this thread.

    I can’t emphasize this FUNDAMENTALLY IMPORTANT point enough. Co2 doesn’t warm the Earth, it just slows down the cooling rate. However, if Miscolczi is correct, and no-one has yet refuted his paper, then the hydrological cycle will speed up to compensate for any increase in co2, and the ‘greenhouse’ will remain in dynamic equilibrium.

    Therefore, as I outlned in my previous post, it’s changes in insolation at the surface which change the global temperature through raising/lowering ocean heat content. The claim that the sun can’t be responsible for late C20th warming is refuted by analysis of the total solar originated energy retained in the oceans over the period of record of ocean heat content, and the calculation of the steric component of sea level rise.

    To sea the truth of this, simply consider the relative heat capacities of the players involved, the sun, the ocean and the atmosphere. The top two meters of ocean have as much heat capacity as the entire atmosphere above. The tail does not wag the dog, as the simple time series plot of SST vs LT proves: Ocean surface warms first, atmosphere responds 2-3 months later. Cause cannot be preceded by effect.

    Nutshell: The sun heats the ocean, the ocean heats the atmosphere, the extra co2 slows down the escape of the heat to space by a tiny bit, but nowhare near enough to account for global warming.

    Nir Shaviv shows us that though the solar cycle might only change TSI by 1.3W/m^2 over the cycle, and on Leif Svalgaards ultra conservative estimates, the total TSI had a smaller change over the C20th, reaching peaks in the late 50′s and 80′s, that change is amplified by cloud albedo changes which make a bigger difference to insolation at the surface. This accounts for far more of the C20th warming than just changes in TSI. By the time we knock of the spurious UHI signal, there isn’t much left for co2, which is mostly along for the ride.

  18. Roy Clark says:

    We have both reached the same conclusions. It is impossible for CO2 to cause any kind of kind of climate change and has not done so, at least since the present composition of the atmosphre was set by ocean photosynthesis about 1 billion years ago.

    R. Clark, Energy and Environment 21(4) 171-200 (2010) ‘A null hypothesis for CO2’

    The paper is behind a paywall. I can e-mail a reprint if you send an e-mail address. I have also posted the discussion on my website at

  19. tallbloke says:

    sphaerica says:
    July 22, 2010 at 3:30 pm (Edit)

    What statistical method did you use to compute the trend lines in your first graph?

    Your description of computing the “Cumulative sunspot total…” is unclear to me. Can you write it mathematically, so that I can be sure that I understand how you are deriving the numbers?

    I just used builtin linear trend function.

    The average sunspot number 1750-2010 is 42. It also happens to be the ocean equilibrium value as near as I can tell. I simply took the difference between 42 and each monthly value, totting up a cumulative value as I went. So if the first month was 62 the first value would be 20. If the second was 36 the running total would become 14, because 32 is 6 less than 42, and so 6 is subtracted from the running total.

  20. sphaerica says:

    Trend lines are very susceptible to selection of end points.

    By using 42 as an average number, what you’ve basically done is to make any values which exceed 42 increase your trend, and any number below 42 decrease your trend. Since sunspot numbers were below 42 until around 1920-1935, and above after, you’d undoubtedly get a smoothly meandering line down until 1935, with a steady meandering rise afterward.

    Being able to find a point where you can match these two in no way shows causation, or even correlation.

  21. sphaerica says:

    To put it another way, in English your premise amounts to:

    “When sun spot counts are below 42 the climate cools, and above 42 the climate warms.”

    This needs a physical reason, an explanation for why 42 is a magic number (outside of Life, the Universe, and Everything) and if possible evidence that this was in fact the case before 1900.

  22. tallbloke says:

    ~42SSN is the long term average over the period of record from 1740 to now. Because we have a mere 260 years of reasonably reliable sunspot numbers, this in itself can’t be assumed to be the very long term average, but it coincides with the average number obtained through analysis of periods during which sea surface temperature neither rises nor falls much. Taken together, it seems reasonable to use the number as the basis for a working hypothesis that it is somewhere close to the ocean equilibrium value.

    The physical reason is that along with the amplifying factor identified by Nir Shaviv in his paper on using the oceans as a calorimeter, small changes in the sun’s output are sufficient to explain the rise and fall in ocean heat content observed. This is further verified by the first part of Roy Clark’s work discussed on this blog.

  23. HR says:


    The nature of your graph of cumulative sunspot number seems to be highly dependant on the choice of the magic number. If you choose a much higher number then many more of the SSN-magic number values become negative and you end up with a downward trend (with bumps) from the start of the time series. If you choose a much lower number then the reverse is true. I tried this out with a monthly mean SSN index (I couldn’t fine an area index online), I don’t expect this difference makes any difference to this general observation.

    Given this fact I think a better justification for the use of 42 is required.

  24. tallbloke says:

    Hi HR, thanks for dropping by.
    All ideas are welcome, as are collaborators in the investigation. One of the problems here is that we are dealing with two variables: The sunspot number (as a proxy for TSI), and albedo (cloud cover levels). Nir Shaviv’s paper ‘Using the oceans as a calorimeter’ is well worth a read.
    If Solar activity and cloud cover are related as Svensmark postulates, then we have a useful method for generating an ocean heat content reconstruction.

    So I’m not sure how we can get exactly ‘the right value’. But this doesn’t worry me much, because in climate science, no-one is sure of the right values for anything – if only they’d admit it. 😉

    My approach has been to find a 40 year period where the ocean temp didn’t change much (according to the HADsst2 reconstruction), and average the sunspot value over that period. Looking at the way the curve flattens around the 1970’s based on the result I think I’m somewhere near right.

    My interest is more in building a useful engineering estimate than in attempting to ‘slam dunk’ the problem.

  25. HR says:

    I pretty much agree with you but the ‘beauty’ of this is the correlation of the two graphs, rather than rely on one of my own favorite instruments ‘HR’s eyeball’ I’d prefer to have things backed up with hard evidence. As you say the likelihood of nailing one driver of the climate is unlikely both because of data quality and the complex reality.

    I got one more request, I tried looking on Web of Science for papers that work with cumulative sunspot data and couldn’t find anything. Do you know of any?

  26. tallbloke says:

    Notwithstanding the small fluctuations in the amount of geothermal energy making its way through the mantle to the surface and seabed, there is only one driver of the climate: the Sun.

    Whether extra amounts of minor trace gases get released by ‘natural’ events or ‘un-natural’?? human activity, all those combustibles were created by the Sun’s energy. If we proceed from there into the climate complexity, we won’t go too far wrong.

    As far as I know, I was the first person to do the integration of solar data. I have seen two other analyses subsequent to my own. One person claimed they got a similar analysis from a colleague, but didn’t provide a document with a date on it.

    The climate science mainstream has studiously avoided the issue of oceanic retention of solar energy, and tries to make out co2 is responsible for the late C20th increase in ocean heat content. This is patently false. They have repeatedly juggled the numbers so it looks like a 1.7W/m^2 forcing, but I calculated from the the steric component of sea level rise in the ’93-2003 decade that there must have been an additional 4W/m^2 going into the ocean. This would be due to raised solar activity and (consequent) lowered albedo.

  27. E.M.Smith says:

    I think you are on to something here.

    Per the “magic number 42”: To the extent that sunspots are a decent proxy for solar output, there will be some average solar output over the interval where it is roughly in heat balance with the present temperature of the planet (where long term temperature equilibrium was reached against average solar output during some interval) and that ought to be reflected in sunspot numbers being in balance as well.

    IFF there is a very long term cycle, then that “magic number” would drift over that very long term period (that is, you would find a new higher or lower earth temperature and a new higher or lower sunspot average number that was ‘in balance’ with that different heat balance point).

    That you find “42” simply means that, at this present more or less average temperature, that’s the point where solar output, or control, functions to hold THAT average temperature.

    I’m fine with that.

    Basically, you’ve found the thermostat right now is set at 72 F and you are watching the gas meter wiggle up and down as the temperature wanders from 68 F to 74 F. No real problem. And if you had 10,000 years of data, you might find it was at 76 F with wobbles of 74 and 80. If 100,000 years, 60 F with wobbles of 56 F and 64 F.

    This analogy would imply that during major transition events there could be a skew or minor decouple (say, as the Milankovich cycle changes the S.H. percent of insolation or as the N.H. glaciates and changes the albedo), but that still does not mitigate the utility of the identified ‘control input function’ over shorter terms (of mere centuries 😉

    So Right Now, with this celestial relationship and this albedo, the set point is matched to 42 on the gas flow meter. Fine. And we watch the gas flow wobble back and forth around 42 and see the room temp wobble back and forth apace. Still fine. Now just go find the air handling system and the house ducts …

    Basically, I’m happy with using a “magic number” that is tautologically derived from the average over the period of the data. I suspect you ought to be able to do a least squares fit through the data, use a midpoint, and get a very similar result.

    The hard bit is then taking all the furnace, ducting, and thermostatic controls apart and showing how they work…

  28. tallbloke says:

    Hi E.M.
    Great! You ‘get it’. I’ve been waiting for someone with your abilities to see the picture I’m trying to paint, and I need your help because my limitations post crash prevent me from following through all the ramifications of this discovery. Please collaborate with me. I’ve just been playing around with outgoing longwave radiation, sunspot numbers and detrended temperature series. It’s looking good. This should improve the ‘fit’ even more, and offer further insights into the way the ocean stores and releases heat on decadal timescales. What I have so far supports my ideas about the ocean going into heat release mode when the sun gets quiet. The OLR swings are around 4W/m^2 over the solar cycle, and this balances the variation in incoming solar multiplied by the terrestrial amplification factor determined empirically by Nir Shaviv.

    Here’s my interim result:
    temp index

    You must read

    And also, you should read Shviv’s as yet unpublished work on spiral arm crossings and long term temp variation.
    to get a handle on the really long term stuff.

  29. tallbloke says:

    Joel Shore says:
    December 27, 2010 at 10:33 am

    tallbloke says:
    Small changes in SST cause bigger changes in atmospheric temperature. You can see this in the effect of the 1998 El Nino on SST and UAH lower troposphere temps. Higher emission by the ocean is to some extent suppressed by higher LT temps.

    Actually, the greater response in lower troposphere temps will tend to limit the effect of the solar forcing and tend to speed the approach back to equilibrium because this greater rise in temperature means (via the Steffan-Boltzmann equation) that there is a larger rise in radiation back out into space. This is why the “lapse rate feedback” is a negative feedback in the climate system.

    I didn’t choose an “arbitrary “neutral” level” for the ocean equilibrium value, I derived it from two different lines of investigation; the long term sunspot average over 250 years, and the average value over an extended period when SST didn’t vary much.

    If you prefer, you can change “arbitrary” to read that it is a fitting parameter that you have essentially adjusted to get agreement with the data. Hence in any comparison of your integrated solar vs the SSTs, one needs to be aware that there are essentially two-parameters in that fit (this one and the one controlling the relative scaling of the SSTs vs the cumulative total sunspot area).

    I understand why this can’t be regarded as a rigorous scientific procedure. It is an engineering estimate. The reason I think it is reasonable to use it is that since our data is so uncertain for SST’s and OHC anyway, it’s as good as anything else out there. You may disagree.

    Well, I think the point is that there are two extremes: one that compares the current solar activity to the temperature and yours that compares the cumulative (i.e., integrated) solar activity to the temperature are two extreme case. The correct picture would be somewhere in between these two extremes since there is a cumulative effect but it is also going to be reduced by the negative feedback that higher temperatures result in higher terrestrial emission back out into space. Essentially, one applies in the limit that the heat capacity in infinitesimally small and the other in the limit that the heat capacity is infinitely large. (One way to incorporate this would be to have the “neutral level” itself be a function of the SSTs since the is effectively what is going to be the case…because as the SSTs rise and the atmospheric temperatures rise also, the amount of heat that the earth radiates back out into space is going to increase.)

    Cheers and seasons greetings to you.

    And, likewise to you!

  30. tallbloke says:

    Yes, I had given some thought as to how much the equilibrium value might alter with changing temperature. I came to the conclusion, having studied the relationship between OLR, TSI and the SOI, that it doesn’t make a big difference. Cloud variation caused by solar activity is a bigger issue.

    Thanks for the feedback.

  31. […] discussion here: LikeBe the first to like this post. […]

  32. Olavi says:

    Abdusamatov has done larger paper about Suns effects on earth climate. There is parts of paper from S. Solank,i Krivova&al

  33. Doug Proctor says:

    The idea that solar energy variations is not an important factor in changes in global temperature is rooted (I think) in the claim that long-term variance is 0.1% of the “av” annual value of 340.5 W/m2 (down 1.15W/m2 recently. Hah! So much for unchangable). On what basis has this high level of consistency been based?

    A simple calculation shows that on the lighted side of the planet, the differnce in insolation July 4 to January 4 (give or take) is 40.8 W/m2. Hemispheric differences due to axial tilt mean that approx 63% falls on one or the other pole depending on which is pointing at the sun (for one day, that is, but still for half the year >50% of the insolation is falling on one hemisphere). Albedo differences due to ground conditions are significant hemispherically. Albedo differences due to cloud cover are time and local dependent and without doubt vary from year to year or even decade to decade. All these variations are averaged out to 340.5 W/2 coming in with a global albedo of about 0.296. What studies show that the time, hemispheric differences and time/location variiations in albedo smooth out to be less than the alleged solar variation of 0.1%?

    Work on 0.25W/m2, or Trenberth’s concern about the “missing” 0.85W/m2 of heating, presuming it to be deep in the oceans (implying that ocean turnover to the 1000m depth is at the annual level to 1W/m2 out of the incoming calculated 288.5 W/m2) assumes that the year-to-year variation in what reaches the planetary surface is at least half of the discussed value, or less than 0.25/0.43W/m2. On what basis is this claim?

    Time, location and albedo changes regionally are significantly greater than the heating levels under discussion. What studies are there that say these time and location impacts don’t add in enough heat to distort the global temperature calculations on a multi-decadal basis?

    The devil seems to me to have a possible hiding place in the assumptions behind global averages. Since small errors may not be random, and may not cancel out in the short-term, but through constructive interference of patterns may be additive until the patterns become destructive, long-term rises and falls can be part of the natural variation. What proof is there that nature is as steady these days as claimed?

  34. […] by making empirical comparisons between SST and the sunspot number (SSN), we find that there is a consistent relationship between the sunspot number and ocean surface temperature when the effects of the oceanic […]

  35. […]… Share this:ShareFacebookDiggTwitterStumbleUponPrintRedditEmailLike this:LikeBe the first to like this post. […]

  36. olsonjs444 says:

    I had performed the identical analysis as tallbloke and achieved the same results back in 2005. This started me blogging against AGW nonsense. As an astrophysicist and radiation expert, with over 30 years of real world work experience, I’m grateful for the opportunity to add my voice to the chorus of those exploring the real mechanisms driving climate change.

    @tallbloke: sunspots have now given way to 10.7cm solar flux due to the remarkable correlation shown here:

    Has anyone converted 1750 – 2011 sunspot numbers to estimates of 10.7cm solar flux? As 10.7cm flux is a regular fraction of total solar flux, this value can be reliably used to estimate total solar energy (and TSI) all the way back to 1750.

  37. tallbloke says:

    Hi Olsonjs and welcome. I’d be very interested to see a copy of your analysis. I have myself created a TSI proxy from sunspot numbers and used it to generate a simple energy model which can be used to predict and reconstruct the Earth’s surface temperature.

  38. Rafael Molina Navas says:

    For DOUG PROCTOR or anybody else who could help me.
    Months ago I sent what follows to Roy Spencer and others at his blog. No answer.
    “When you say that some forcing or the like is of, f.e. 0.5 w/m2, energy per unit of time and unit of surface, what surface and time is supposed to be considered?
    In other words, how many m2 and hours, multiplied by the given w/m2 figure, would give us the total energy in watt*hour, let us say per day?
    Because this is something not uniform: some radiations are related to only the effective real surface of CLOUDS and their real existing day time, others are related to the surface of maximum Earth transverse section, 24 h. a day … and others even to the total surface of the Earth, 24 h. a day … And what is important is total energy per day, not power per m2”.
    Thank you.

  39. olsonjs444 says:

    I’ve corresponded directly with Dr. Spencer. The calculations you’re seeking are well-understood, and can be reproduced by studying the material at this site:

    And the following pages:

    Of course, you will need to throw out the false concept of a “solar constant” which is used on these “simple” pages. There is nothing constant about the level of solar energy being thrown out into the solar system, let alone arriving at the top of Earth’s atmosphere on a day-to-day basis. Annual solar radiation more closely approximates a constant, but even this varies over the course of a solar cycle, and from cycle to cycle (by 0.1% per cycle or more… we only have three half-cycles worth of direct observational data).

  40. Stephen Wilde says:

    “The Death Blow To Anthropogenic Global Warming” June 4th 2008

    “It is true that, as the alarmists say, since 1961 the average level of TSI has been approximately level if one averages out the peaks and troughs from solar cycles 19 through to 23.

    However, those solar cycles show substantially higher levels of TSI than have ever previously occurred in the historical record.

    Because of the height of the TSI level one cannot simply ignore it as the IPCC and the modellers have done.

    The critical issue is that having achieved such high levels of TSI by 1961 the sun was already producing more heat than was required to maintain a stable Earth temperature. On that basis alone the theory of AGW cannot be sustained and should now die.

    Throughout the period 1961 to about 2001, there was a steady cumulative net warming effect from the sun. The fact that the TSI was, on average, level during that period is entirely irrelevant and misleading.

    It is hardly likely that such a high level of TSI compared to historical levels is going to have no effect at all on global temperature changes and indeed during most of that period there was an enhanced period of positive Pacific Decadal Oscillation that imparted increasing warmth to the atmosphere. My link below to article 1041 contains details of my view that the sun drives the various oceanic oscillations which in turn drive global temperature variations with all other influences including CO2 being minor and often cancelling themselves out leaving the solar/oceanic driver supreme.”

  41. tallbloke says:

    Heh, good stuff Stephen. We’d been thinking along the same lines before we read each other’s stuff.

    You reasoned out the situation before I got around to quantifying it.

  42. olsonjs444 says:

    One last thought about sunspot numbers: add together the monthly sunspot numbers over each sunspot cycle from minimum to minimum. You’ll find the sum of sunspot numbers for the early C20th are only about one-half the sum of sunspot numbers for the late C20th. If we view these integrations as proxies for the amount of solar energy added to the earth’s climate system, then the first decade of the C21st should have set records (new high temperatures).

    However, it is interesting to note that not one single U. S. state record high temperature has fallen over this period of “unprecedented” warmth. If the total area of the U.S. comprised more than a few percent of the earth’s surface, then this might be significant. Instead, I think it is wise to remain skeptical until weather monitoring stations can measure temperature accurately from -100 C to +60 C.

  43. […] the OHC data better as a forcing. To build this I first identified the value at which the ocean neither gains nor loses energy from a level period of SST in the C19th, before co2 became an issue. Then I made a cumulative […]

  44. scf says:

    Very interesting stuff. Great blog.